binve (< 20)

binve's CAPS Blog

Recs

9

October 11, 2011 – Comments (7)

I personally am getting pretty tired of reading inconsistent economic and macroeconomic 'analysis'. The classic meaningless ratio that gets thrown around all the time (a la Reinhart and Rogoff) is National Debt / National GDP. The National Debt is a stock (number of dollars), The National GDP is a flow (number of dollars per time, such as quarters or years).

As I discuss here: Why Deficit Spending and Creative Destruction are not Mutually Exclusive Positions

A 'flow' is an economic quantity (in this post it is assumed that the quantity is in terms of US Dollars) measured over a unit of time. For example GDP is a flow (the net economic activity generated in a quarter or a year). The US Budget deficit is a flow (how much the Federal Government spends more than it taxes over a year).

A 'stock' is an economic quantity (again in terms of Dollars). The US National Debt is a stock.

'Flows' accumulate to 'Stocks'. At the end of a measurement period, the US Budget Deficit accumulates to the National Debt.

You cannot compare stocks and flows in any meaningful way. You compare stocks against stocks, you compare flows against flows. These both give ratios. But comparing stocks and flows gives you a number in units of time. For example: The US National Debt (say \$50 Trillion) / The US GDP (say \$10 Trillion/yr) gives you a result of 5 years. This is not a meaningful metric, because it has no context.

When 'anlaysts' start ratioing stocks and flows, such as David Brooks does in the article below, it pays to put on a A Decent Macroeconomic Filter

----------------------------------

Bathtubs for Beginners
By James Kwak

http://baselinescenario.com/2011/10/11/bathtubs-for-beginners/

In economics life there’s a basic conceptual distinction between a flow and a stock. A flow is a something that occurs over some period of time, like water pouring from a faucet into a bathtub. A stock is something that exists at a specific moment of time, like the water in that bathtub. You measure a flow over a period of time (e.g., gallons per minute); you measure a stock at a specific moment in time (e.g., gallons). For a business, the income statement (revenues and costs in a year) measures flows, while the balance sheet (assets and liabilities) measures a stock. That’s why the income statement is dated for a year (or a quarter) and the balance sheet is dated for a specific day. Everyone understands this. If you didn’t, you would get confused between your salary and your bank account.

But not David Brooks.

In today’s self-indulgently contrarian column, Brooks argues that the Occupy Wall Street movement is made up of “small thinkers.” Here’s his evidence:

“They will have no realistic proposal to reduce the debt or sustain the welfare state. Even if you tax away 50 percent of the income of those making between \$1 million and \$10 million, you only reduce the national debt by 1 percent, according to the Tax Foundation. If you confiscate all the income of those making more than \$10 million, you reduce the debt by 2 percent. You would still be nibbling only meekly around the edges.”

This is incoherent to begin with. Tax policy directly affects flows, not stocks, so its impact on the national debt (a stock) is indeterminate unless you specify a length of time for the policy to be in place.

#1) On October 11, 2011 at 11:20 AM, PeteysTired (< 20) wrote:

Can we assume the author meant that the time period is a year?

Report this comment
#2) On October 11, 2011 at 11:53 AM, davejh23 (< 20) wrote:

"Even if you tax away 50 percent of the income of those making between \$1 million and \$10 million, you only reduce the national debt by 1 percent"

In this case, I think it's safe to assume that they're talking about a 50% tax rate on top income earners over a year.  Of course, the 1% reduction in the \$14+ trillion national debt must depend on several other assumptions as well.

Based on your flow/stock argument, maybe they should discuss confiscating 50% of wealth instead of income.  I saw an article recently that pointed out that the combined wealth of the wealthiest 1% is only, I believe, \$1.X trillion...10-15% of the national debt.

Report this comment
#3) On October 11, 2011 at 12:42 PM, wolfman225 (75.60) wrote:

Don't forget to allow for the effects of any potential consequences of specific tax policies.  The situation isn't static, despite the claims of some "tax the rich" proponents that a tax rate of x% will yield an increase in revenue of \$XB dollars over a specified time frame (usually 10yrs).  The economy is dynamic and one predictable result of increased tax rates is the sheltering of income, which usually results in the decrease in actual revenues.  Just as you can't compare flows/stocks directly, any ratio that doesn't take into account the dynamics of future policies and reactions is likewise of little value.

Report this comment
#4) On October 11, 2011 at 12:50 PM, ikkyu2 (98.89) wrote:

Stock/flow, as you have defined it, is a perfectly meaningful metric.  The question is, "If the entire annual GDP of the USA went to pay down the national debt, how long would it take to pay it off?"  You get an answer in years (or seconds, or quarters, or whatever time units you like.)

People compare "stocks" to "flows" all the time.  The most common measure is certainly the price/earnings ratio.

Report this comment
#5) On October 11, 2011 at 1:26 PM, binve (< 20) wrote:

>> Stock/flow, as you have defined it, is a perfectly meaningful metric.  The question is, "If the entire annual GDP of the USA went to pay down the national debt, how long would it take to pay it off?"  You get an answer in years (or seconds, or quarters, or whatever time units you like.)

100% disagreed. Yes, you can come up some sort of explanation like that, but is it meaningful? No. Will the population forego all consumption, investment, savings, etc for years to pay off a national debt (which is nothing more than the accumalted non-government sector savings to begin with)? No. So to suggest this is a meaningful way to interpret this meaningless metic is wrong.

If I have a block of metal at it has 10000 Joules (stock) of stored thermal energy in it and I have a thermal sink that pulls out 100 W (which is Joules/sec which is a flow), then it takes 100 sec to remove all the energy, right? Nope. That metric neglects capacitance, energy distribution within the metal, the ability of energy to redistribute through the metal, discontinuites and internal geometry, etc.

So the 1000 J / 100 W = 100 s metic is just as meaningless as a 10 Trillion \$ / 1 Trillion \$/year = 1 year metric. It does nothing to characterize the system in question.  As a standalone number it is meaningless.

Japan is soverign issuer of its own currency. A debt/gdp of 200% is no more or less serviceable than a debt/gdp of 1000% or a debt/gdp of 0%.

That is very different than Greece, which is *not* soverign issuer of the Euro and faces very differnt constraints.

This again illustrates why debt/gdp is a useless and highly misleading standalone metric.

>>People compare "stocks" to "flows" all the time...

They do. It doesn't mean its right, and in fact most of the time it isn't

>> The most common measure is certainly the price/earnings ratio.

LOL! You have got to be kidding me!

Okay, the P/E of a company is 8. Is this good or bad?

They answer is: you have no idea.

What sector is  the company in? What are the reliability of earnings? Is the company going bankrupt (good earnings 4 quarters ago, but negative earnings in the last 3 quarters) but the TTM earnings is still a positive number? Can the company use resources to increase earnings easily?

P/E as a standalone number tells you absolutely nothing about the underlying value of the company.

As such it is also a highly misleading value.

Report this comment
#6) On October 11, 2011 at 2:00 PM, Frankydontfailme (29.44) wrote:

Well binve, we agree that P/E is pretty much meaningless :)

You should add the caveat: any debt to gdp of a nation which issues its own currency is servicable as long as it is not largely dependent on imports, and as long as exporters accept their currency. (also, there's the issue of interest rates, which I believe you've explained in previous posts, but your explanation nonetheless does not have any relevance to the real world where interest rates DO MATTER)

Report this comment
#7) On October 12, 2011 at 3:03 AM, ikkyu2 (98.89) wrote:

If I have a block of metal at it has 10000 Joules (stock) of stored thermal energy in it and I have a thermal sink that pulls out 100 W (which is Joules/sec which is a flow), then it takes 100 sec to remove all the energy, right? Nope. That metric neglects capacitance, energy distribution within the metal, the ability of energy to redistribute through the metal, discontinuites and internal geometry, etc.

Agreed that this statistic alone is not interesting.  But if the answer is 100 sec in February, 80 sec in August, and 110 sec in October, a real scientist would find that intriguing.

Report this comment